US2024054801A1PendingUtilityA1

Arrangements for digital marking and reading of items, useful in recycling

Assignee: DIGIMARC CORPPriority: Mar 26, 2020Filed: Aug 28, 2023Published: Feb 15, 2024
Est. expiryMar 26, 2040(~13.7 yrs left)· nominal 20-yr term from priority
G06V 30/224G06T 7/90B07C 5/3416B07C 5/3422B65G 47/493G05B 13/027G06T 1/0014G06T 1/0021G06T 7/0004H04N 7/18G06F 18/21G06F 18/24H04N 23/56H04N 23/73G06V 10/811G06V 10/82G06V 10/454B07C 2501/0054G06T 2207/10024G06T 2207/10048G06T 2207/20084G06T 2207/30108G06V 10/58G06V 2201/06G06F 18/256B29B 17/02B29B 2017/0203B29B 2017/0279B29B 2017/0282B29B 17/0412
74
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

Images depicting items in a waste flow on a conveyor belt are provided to two analysis systems. The first system processes images to decode digital watermark payload data found on certain of the items (e.g., plastic containers). This payload data is used to look up corresponding attribute metadata for the items in a database, such as the type of plastic in each item, and whether the item was used as a food container or not. The second analysis system can be a spectroscopy system that determines the type of plastic in each item by its absorption characteristics. When the two systems conflict in identifying the plastic type, a sorting logic processor applies a rule set to arbitrate the conflict and determine which plastic type is most likely. The item is then sorted into one of several different bins depending on a combination of the final plastic identification, and whether the item was used as a food container or not. A variety of other features and arrangements are also detailed.

Claims

exact text as granted — not AI-modified
1 - 13 . (canceled) 
     
     
         14 . A method employing first and second image processing systems that operate on imagery captured by one or more cameras viewing a waste stream on a conveyor belt, the first system comprising a convolutional neural network classification system, the second system comprising a watermark detection system, the method including: the convolutional neural network classification system classifying a first item on the conveyor belt and providing data to the watermark detection system including location information for the first item, and the watermark detection system responding to said data by not attempting a watermark reading operation on image data corresponding to said location information. 
     
     
         15 - 36 . (canceled) 
     
     
         37 . A method comprising the acts:
 capturing imagery depicting a conveyor belt conveying items thereon;   from said captured imagery, producing data indicating appearance of the conveyor belt with no item thereon.   
     
     
         38 . The method of  claim 37  in which the conveyor belt is a looped conveyor belt and the method includes:
 capturing frames of image data depicting portions of the conveyor belt conveying said items as the belt moves past an imaging system in a waste processing facility; and 
 analyzing a captured frame of image data to identify an empty region of the belt, said analyzing includes performing a correlation operation between first image data from the captured frame, and image data earlier-gathered, to identify said empty region of the conveyor belt. 
 
     
     
         39 . The method of  claim 38  that further includes generating a match metric, and comparing said match metric against a threshold to identify said empty region of the belt, the match metric comprising a correlation value produced from said correlation operation. 
     
     
         40 . The method of  claim 38  that includes performing a correlation operation at each of a plurality of spatial alignments between the first image data and the image data earlier-gathered, yielding a set of correlation values, each associated with a respective spatial alignment, and computing said match metric by determining a peak value among said set of correlation values. 
     
     
         41 . The method of  claim 40  that includes computing said match metric as a combination of (a) said peak value among said set of correlation values, said peak value being associated with a first spatial alignment, and (b) a second correlation value, the second correlation value being associated with a spatial alignment that is adjacent to said first spatial alignment. 
     
     
         42 . The method of  claim 41  in which said combination comprises a weighted sum. 
     
     
         43 . The method of  38  in which said analyzing includes subtracting a set of fixed pattern noise from the first image data prior to performing the correlation operation. 
     
     
         44 . The method of  claim 38  that further includes
 assembling a patchwork collection of image excerpts depicting empty regions of the conveyor belt, to thereby produce data indicating appearance of the conveyor belt with no item thereon. 
 
     
     
         45 - 48 . (canceled) 
     
     
         49 . A method comprising the acts:
 capturing first imagery depicting waste material, including an item, on a conveyor;   capturing second imagery depicting waste, including said item, on the conveyor;   determining that the item is moving at a different rate than said conveyor; and   operating a diverter to remove the item from the waste on the conveyor, taking into account said moving at a different rate.   
     
     
         50 - 57 . (canceled)

Join the waitlist — get patent alerts

Track US2024054801A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.